# Strukora

> "Structure before software."

> Diagnostic-first AI consultancy. We audit business processes, identify high-ROI AI
> opportunities, prove the numbers before anything is built, and say so when the answer
> is not AI.

Strukora works with mid-market companies, roughly 200 to 1,000 employees, that have
already spent on AI without a measurable result. The entry product is a fixed-scope,
fixed-price AI audit. Implementation follows only for opportunities the audit justifies.

## How to engage
1. Free 7-question readiness assessment. No email required for the summary.
2. Free 30-minute opportunity review. A real assessment of where you are, not a sales call.
3. Paid audit. Scope, duration and deliverable are published; the price is set on the call,
   because it depends on how many processes are in scope and how scattered the data is.

## Positioning
- We publish the scope, duration and deliverable of every engagement.
- Step 04 of every audit can conclude "build nothing". That is a valid outcome.
- We start with data readiness, not model selection.
- Our technical differentiator is identity and permission mapping in retrieval systems:
  how documents inherit access rights, and how role hierarchies map to sensitivity.

## Services
- /services/ai-audit-strategy/ — diagnosis and opportunity scoring
- /services/rag-knowledge-systems/ — permission-aware retrieval over company documents
- /services/ai-agents/ — narrowly scoped agents with measurable success criteria
- /services/ai-automation/ — deterministic automation, often the honest answer
- /services/data-ai-engineering/ — the data layer everything else depends on

## Solutions by problem
- /solutions/knowledge-management/ — teams re-finding information that already exists
- /solutions/document-processing/ — manual extraction, entry and verification
- /solutions/customer-support/ — high volume of near-identical requests
- /solutions/operations/ — repetitive internal workflows
- /solutions/sales/ — research and follow-up overhead
- /solutions/hr/ — screening and onboarding admin

## Free tools
- /tools/ai-readiness-score/ — LIVE. 7-question assessment, scored by an explicit rules
  engine rather than a language model, so the result is reproducible and auditable. The
  score and the top three opportunities are shown on screen with no email required.
  Notable behaviour: if the answers indicate the underlying data is scattered or unknown,
  the top recommendation is to fix the data foundation, not to build an AI system.
- /tools/ai-roi-calculator/ — in progress
- /tools/rag-cost-calculator/ — in progress
- /tools/ai-project-cost-estimator/ — in progress

## Site map
Live: / · /services/ and 5 service pages · /solutions/ and 6 solution pages · /ai-audit/ ·
/book-a-review/ · /how-we-work/ · /about/ · /contact/ · /tools/ · /tools/ai-readiness-score/
Not built yet: /work/ (labs, benchmarks, case studies) · /resources/ (guides, research,
comparisons) · /blog/

## Evidence
- /work/labs/ — internal engineering write-ups, explicitly labelled experimental research
- /work/benchmarks/ — measured benchmarks with published methodology and downloadable raw data
- /work/case-studies/ — client work, only with client agreement

## Editorial rules we hold ourselves to
- No invented clients, logos or testimonials.
- No metric published without the source and what that source actually measures.
- No benchmark figure that was not measured.
- Model prices live in a dated file and are never quoted from memory.
- All experimental work is labelled as such.

## Contact
- /contact/
- /book-an-ai-audit/